[Re] FairDICE: A Fair Tradeoff in Multi-objective Offline RL

ArXi:2603.03454v2 Announce Type: replace Offline Reinforcement Learning (RL) is an emerging field of RL in which policies are learned solely from nstrations. Within offline RL, some environments involve balancing multiple objectives, but existing multi-objective offline RL algorithms do not provide an efficient way to find a fair compromise. FairDICE (see arXi:2506.08062v2) seeks to fill this gap by adapting OptiDICE (an offline RL algorithm) to automatically learn weights for multiple objectives to e.g. incentivise fairness among objectives.